Conversion - why the winner takes all

Your website conversion rate will go a long way to decide your success or failure online.

This is because internet marketing can be a bidding war. And the site with the highest visitor value can bid highest and often ends up owning all the traffic.

(visitor value = your average profit per visitor)

The easiest way to raise your visitor value is through getting a higher percentage of your visitors to become paying customers. And the best way to do that is through something called "split-testing".

Split-testing allows you to improve the effectiveness of your site scientifically. Here's how it works:

(1) You choose a key page on your website

(2) You create a new version of that page

(3) You use software to run the two versions of the page against each other i.e. half the people see the old version, half the people see the new version

(4) The split-test software tracks how many sales/responses come from each version

(5) Once you have enough data, you can see which page is more effective. You keep that version and delete the "loser"

(6) You do another split-test

How big can these improvements be?

If you'd like to see the results of a split-test I did for a client, click here: 28.9% growth in a month.

Want another? Take a look at this screenshot of an ongoing (at time of writing) test that is showing a 32.3% increase in sales from the main product page - and all I did was re-write the bullet points!

And another where the increase produced an improvement of between 21.5% and 53.5% over the original version of the page

I say "between 21.5% and 53.5%" because of the margin for error. If I had run the test for longer, I would be able to give you a more accurate number. However, the point of split testing is to get a winner, delete the loser and then move on to the next test.

However, let's take the lowest amount, that's a 21.5% increase in sign-ups for that website. Imagine you were to increase your online conversion, how much extra money would that mean for you? Per month? Over a year? Or over 5 years?

Getting 20-33% Increase on Your Website

So, you've seen a number of examples of tests I've run and you've seen the sorts of increases in sales these tests produce.

If you're interested in getting the most out of your hard-earned website traffic, I offer a low-cost split-testing service. I'll take a look at your site, identify a key page that's open to likely improvement and then create a new version to test.

This service costs just £97. If that sounds like a lot, ask yourself, what percentage increase would you need to get to make an extra £97 a year from your site? 1%? Less?)

Or, if you prefer, I'd happy to take the all the risk for you. If I can't increase conversion by at least 10%, you pay me nothing. However, in that case, the price goes up to £179.

Again, ask yourself, how much would you make if you increased your sales by 10%?

... And that's just one test. We could be testing lots of things on your site and finding 10-20% increases all over the place...

Here's the bad news

This offer comes with 3 conditions:

(1) As we can only test what we can measure online, this split testing only works for websites that are intended to create online conversions e.g. completing an order form, filling in an enquiry form, clicking a particular link, signing up to a newsletter etc.

If your website is designed to get people to call you on the phone or to visit your shop, this won't work.

(2) You need to have enough traffic to generate reliable results.

If you put a handful of visitors through a split-test, the results are completely unreliable. So, for me to split test a page on your site, I'd want it to have at least 25 conversions per month.

(3) You have to be able to change your site.

If you can't change your site (or get your webmaster to do it for you), we can't set up a split test. It's as simple as that.

However, if you've got a site that can be tracked, that gets a decent number of conversions and can be changed, you can benefit from split-testing.

So, if you'd be interested in me running a test like this for you, email me at steve@stevegibsonconsulting.co.uk and I'll have a look at your site to see if I think I can improve it.

Steve Gibson

PS If you're asking yourself how I do these tests, I use a piece of software from google called "Website Optimizer". This tracks the visitors and conversions from each version of the page and creates a set of results with percentages. All the screenshots you saw in the examples I gave were taken from Website Optimizer.